Activity, Peroxide Compound Formation, and Heme d Synthesis in Escherichia coli HPII Catalase

Transcription

1 ARCHIVES OF BIOCHEMISTRY AND BIOPHYSICS Vol. 342, No. 1, June 1, pp , 1997 Article No. BB Activity, Peroxide Compound Formation, and Heme d Synthesis in Escherichia coli HPII Catalase C. Obinger,* M. Maj,,1 P. Nicholls,,2 and P. Loewen *Institut für Chemie, Universität für Bodenkultur, Muthgasse 18, A-1190 Wien, Austria; Department of Biological Sciences, Brock University, St. Catharines, Ontario, L2S 3A1, Canada; and Department of Microbiology, University of Manitoba, Winnipeg, Manitoba, R3T 2N2, Canada Received December 17, 1996, and in revised form March 6, 1997 drogen peroxide but four to five orders of magnitude Wild-type Escherichia coli HPII catalase (heme d more slowly than do hydroperoxidases (1). This peroxcontaining) has 15% the activity of beef liver enzyme ide reactivity is controlled by the protein (2). A suitably per heme. The rate constant for compound I formation positioned distal histidine (also present in myoglobin with H 2 O 2 is M 01 s 01. HPII compound I reacts and hemoglobin) and a second distal residue (arginine with H 2 O 2 to form O 2 with a rate constant of in peroxidases, asparagine in catalases) are important M 01 s 01. Forty percent of HPII hemes are in the com- for compound I formation and heterolytic peroxide pound I state during turnover. Compound I is reduced cleavage. by ethanol and formate at rates of 5 and 13 M 01 s 01 (ph The heme iron in these proteins is usually oxidized 7.0), respectively. Incubation of HPII compound I with by hydrogen peroxide to a ferryl [FeIV O] form. The ferrocyanide and ascorbate does not form a compound second oxidizing equivalent produces either a protein II species. Mutation of His128 to alanine or asparagine radical as in metmyoglobin (3 5), hemoglobin (6), gives inactive protoheme proteins. Mutation of Asnand yeast cytochrome c peroxidase (7, 8) or a porphy- 201 gives partially active heme d forms. Asn201Ala has rin radical as in plant peroxidases (9) and eukaryotic 24%, Asn201Asp 10%, and Asn201Gln 0.4% of wild-type catalases (10). Thus, the reaction of beef liver catalase activity. Asn201His contains protoheme when isolated and converts this via protoheme compound I to a heme with H 2 O 2 produces ferryl iron plus porphyrin p-cation d species. Both distal heme cavity residues His128 and radical or compound I. This is reduced in a single two- Asn201 are implicated in catalytic activity, compound electron step to the ferric state or by a one-electron I formation, and in situ heme d biosynthesis. HPII reaction to give the inactive peroxide compound II (11), Asn201, like the corresponding residue in protoheme in which the iron remains ferryl but the porphyrin radicatalases, may promote H / transfer to His128 imidaz- cal has been lost. ole, facilitating (i) peroxide anion binding to heme and As originally shown by Kirkman and co-workers, (ii) stabilization of a transition state for heterolytic mammalian catalases (12), yeast catalases A and T cleavage of the O O bond Academic Press (13), and the catalases of some prokaryotes (12, 14, Key Words: HPII; peroxide; catalatic reaction; cata- 15) all bind one NADPH molecule per subunit, whose lase peroxide compound; heme transformation; proto- function appears to be protection against hydrogen perheme; heme d; site-directed mutagenesis; E. coli cata- oxide inactivation and prevention of compound II forlase; reaction rates; heme pocket; distal residues. mation. The major catalase from Escherichia coli, HPII, 3 is evolutionarily homologous with the eukaryotic enzyme Hydroperoxidases, the enzymes that metabolize hyprotoheme as a prosthetic group. HPII is a monofunc- and with enzymes from other prokaryotes that contain drogen peroxide, include the catalases and peroxidases. Metmyoglobin and methemoglobin also react with hy- 3 Abbreviations used: HPII, hydroperoxidase II (catalase) of E. 1 Present address: Department of Biochemistry, McMaster Univer- coli; WT, HPII wild type; CatPHFeIII(HOH), ferricatalase; sity, Hamilton, Ontario, L8N 3Z5, Canada. CatPH / FeIV O, catalase compound I; CatPHFeIV O, catalase 2 To whom correspondence should be addressed. Fax: compound II; catalatic reaction, the process of dismutation of pnicholl@spartan.ac.brocku.ca. 2H 2 O 2 to O 2 and 2H 2 O /97 $25.00 Copyright 1997 by Academic Press All rights of reproduction in any form reserved.

2 HEME d SYNTHESIS IN Escherichia coli HPII CATALASE 59 tional catalase with one heme d per subunit (M r 84,200) t 1/2 is the time taken for the absorbance level of H 2 O 2 to decay half associated in a tetrameric structure (16). The heme d its initial value. Overall rate constants were calculated as k Å k / [catalase hematin]. is a doubly cis-hydroxylated form (17) whose pyrrole (b) Steady-state spectral changes. During the steady-state catalring III has been modified by the hydroxyl group ysis of the substrate the ratio of free enzyme [CatPHFeIII(HOH)] to attached to the carbon atom carrying the propionate compound I [CatPH / FeIV O] is a constant determined by the rate side chain being linked to the latter via a lactone bridge constants k 1 (app) and k 2 (app) according to Eqs. [1] [3]: (18) (see Fig. 7A under Discussion). HPII catalase does not bind NAD(P)H (13). [CatPHFeIII(HOH)]/[CatPH / FeIV O] Å k 2 (app)/k 1 (app) [1] The crystal structure of HPII (16) places His128 just above pyrrole ring III (the hydroxylation site) and as we have Asn201 in close proximity (see Fig. 7B under Discussion). These residues are counterparts of His74 and Asn147 in beef liver catalase, which have been postulated k (app) 1 CatPHFeIII(HOH) / H 2 O 2 r CatPH / FeIV O / H 2 O [2] to play a role in H 2 O 2 dismutation (19). HPII CatPH / k (app) 2 FeIV O / H 2 O 2 r CatPHFeIII(HOH) / O2 / H 2 O also catalyzes heme d formation from protoheme in situ using hydrogen peroxide as oxidant (18). Site-directed [3] mutagenesis showed that the distal His128 is abso- From the steady-state concentration of compound I and the mealutely required for heme conversion. sured kinetics of peroxide compound formation, k 2 (app) can be calculated We investigated the role of two distal heme cavity (Eq. [1]). Alternatively, the rate of combination of catalase com- residues, histidine 128 and asparagine 201, in catalytic pound I and hydrogen peroxide may be obtained from Eq. [4], where k is the overall rate constant for the catalatic reaction, p is the activity, in compound I formation, and in heme d synsteady-state concentration of compound I, and e is total enzyme hethesis, and we describe here the behavior both of wild- matin concentration (cf. Ref. 24): type HPII and of the mutants His128Ala, His128Asn, Asn201Ala, Asn201Asp, Asn201His, and Asn201Gln. This paper also reports the rates of compound I formation k 2 (app) Å k /2(p/e). [4] with hydrogen peroxide and of compound I reduc- In the catalatic steady state the Soret band of beef liver catalase tion by ethanol and formate, compares the reactivities compound I has a 12 15% lower intensity than that of the native of protoheme and heme d forms, and proposes that enzyme. If 100% compound I is prepared using peracetate or an alkyl peroxide, the compound I Soret band is only 50% that of the ferric transformation of protoheme to heme d occurs via a enzyme. The steady-state compound I concentration (p/e) is thus 25 protoheme compound I. A previous study characterized 30% and k 2 (app) is therefore three times k 1 (app). the binding and inhibitory action of cyanide on free To evaluate the corresponding steady state of HPII and to calculate enzyme and some of the same mutants (20). values of k 2 (app) and k 1 (app) (Eqs. [1] and [4]), enzyme samples (hematin concentrations 3 to 5 mm) were titrated with hydrogen peroxide. MATERIALS AND METHODS In some cases the proportion of free catalase remaining in the system was determined by the cyanide back titration method of Chance (25). These measurements were carried out with a diode- Materials array spectrophotometer (Beckman Instruments DU7400 uv/vis). Escherichia coli HPII catalases were isolated in the Department (c) Transient-state kinetics. The sequential stopped-flow apparaof Microbiology, University of Manitoba, according to Loewen and tus and the associated computer system were from Applied Pho- Switala (17). Site-directed mutants were prepared as described by tophysics (UK): SX-17MV. In the conventional stopped-flow mode Loewen et al. (18). Concentrations of HPII were estimated using an the apparatus uses two syringes, delivering a total of 0.1 ml/shot extinction coefficient per hematin (21) of 118 mm 01 cm 01 at 405 nm. into a flow cell with a 1-cm light path. The fastest time for mixing The lyophilized protein was dissolved in potassium phosphate buffer two solutions was of the order of 1.5 ms. Each trace consisted of 400 and centrifuged to remove insoluble material. Fisher Scientific sup- data points. The enzyme concentrations for the kinetic experiments plied the 30% hydrogen peroxide. KH 2 PO 4, KCN, and cysteine were were 0.8 and 1.6 mm HPII and the H 2 O 2 concentrations were at least products of BDH Chemicals. K 2 HPO 4 and 95% ethanol were from 10 times in excess to ensure first-order kinetics. Temperature was Caledon Laboratories. Potassium ferrocyanide and sodium ascorbate set to 25 C. Three determinations of rate constants were performed (BDH Chemicals) were prepared as 10 mm stock solutions in distilled for every substrate concentration and the mean value was used in water, kept frozen until use. calculation of the second-order rate constants. Second-order rate constants were calculated from the slope of the line defined by a plot of k obs versus substrate concentration. Methods The reaction of HPII with hydrogen peroxide, k 1 (app), was followed (a) Steady-state catalase assay. The dismutation of hydrogen at 405 nm. Reactions of compound I (produced by incubation of ferricperoxide was followed spectrophotometrically (22) using an extinc- atalase with hydrogen peroxide) with either ethanol or formate [the tion coefficient for hydrogen peroxide at 240 nm of 39.4 M 01 cm 01 peroxidatic reaction of Keilin and Hartree (26, 27)] were measured (23). Measurements were carried out with an Aminco DW-2 spectro- at 405 nm using the sequential stopped-flow method. The delay time photometer linked to a Compaq 286 device with Olis software and ( ms) in the aging loop was calculated from the reaction of hardware. Data were analyzed using the Olis fitting routines for the free enzyme with H 2 O 2 at [H 2 O 2 ] Å 2 5 times [hematin]. Ethanol various types of exponential reaction. HPII and mutant concentrations and formate concentrations were mm. were in the range of 1 10 nm. Pseudo-first-order rate constants All experiments were performed at 25 Cin50mMpotassium phos- for the samples examined were calculated as k Å ln 2/t 1/2, where phate buffer, ph 7.4.

3 60 OBINGER ET AL. FIG. 1. Absolute and peroxide difference spectra of wild-type and mutant HPII catalases. (A) Absolute spectra of wild-type enzyme (4.2 mm) and the mutants N201D (4.0 mm) and N201H (3.9 mm) in50mmpotassium phosphate buffer, ph 7.4, at 25 C. (B) Difference spectra of wild-type HPII peroxide complexes. Conditions as in (A). Hydrogen peroxide was added discontinuously to the resting enzyme [1.4 mm (a), 2.8 mm (b), 4.3 mm (c), and 5.7 mm (d)] and steady-state spectra were recorded immediately after addition using the Beckman 7400 diode array spectrophotometer. (d) Molecular visualization. The structures of the HPII wild-type absorption bands displaced about 90 nm toward the red and mutant enzymes were visualized with a Silicon Graphics Iris end of the spectrum compared to those characteristic of (Unix-based) computer using crystal structure coordinates as in Bravo et al. (16) and the molecular modeling program Quanta. protohematin enzymes, indicating a d type heme as prosthetic group. Mutation of the distal His128 to Ala RESULTS or Asn gave inactive enzymes containing mainly protoheme, whereas substitution of the Asn201 residue by Ultraviolet Visible Spectra Ala, Asp, or Gln was accompanied by retention of the Figure 1A shows the spectra of wild-type HPII and red-shifted spectra. Of the Asn201 mutants only the mutants of interest. Wild-type HPII has a set of Asn201His showed a typical protohematin spectrum.

4 HEME d SYNTHESIS IN Escherichia coli HPII CATALASE 61 FIG. 2. Absolute spectra of wild-type HPII catalase and its primary hydrogen peroxide compound. HPII (wild-type) enzyme (Ç4 mm hematin) was treated with excess (ú1 mm)h 2 O 2 at ph 7 and 25 C. When oxygen evolution had ceased (i.e., free [H 2 O 2 ] r 0), potassium cyanide was added and the proportion of free enzyme present was calculated from a parallel experiment in the absence of cyanide (see Methods and Ref. 25). The expected spectrum of 100% compound I was then calculated and displayed. Other conditions are as in Fig. 1. Titration of wild-type enzyme with hydrogen perox- absorbance at 405 nm of between 10 and 20%, indicating ide resulted in a maximal Ç15 20% decrease in absorbance a steady-state compound I concentration of between (hypochromicity) at both 405 and 590 nm 20 and 40%. With Asn201His, titration with hydrogen (compound I formation). Figure 1B shows typical difference peroxide leads to a maximum hypochromicity of 10 spectra (steady-state compound I minus free en- 15% at 405 nm which is then followed by peak shifts zyme) which contain isosbestic points between native (indicating heme conversion; see below). enzyme and compound I at 367 and 433 nm. These hypochromicities fit a steady-state compound I concentration between 30 and 40% of total hematin, assuming TABLE 1 that 100% compound I formation would give a Soret Kinetic Constants for the Catalatic Reaction of HPII absorbance Ç50% that of the ferric enzyme. This was Wild-Type and Mutant Catalases confirmed by experiments using the cyanide back titration method of Chance (25), results of which are illusk % k cat Steady-state trated in Fig. 2. This shows the calculated absolute Catalase (M 01 s 01 ) WT Heme (s 01 ) comp. I (%) spectrum of HPII peroxide compound I obtained by di- HPII wild type d 26, viding the difference spectrum by the fractional occu- H128N 0.1 b pancy and then adding this difference to the spectrum H128A 0.1 b of the original enzyme. Fractional occupancy ( p/e) was N201 A d 9, N201D d 7, N201 Q d ND ND determined by the proportion of cyanide complex (20) formed by the steady-state mixture compared to that N201H protoheme b ND formed in the absence of peroxide (1 0 p/e). N201H heme d cis d ND ND The results of similar experiments with HPII mu- Note. Conditions: 50 mm potassium phosphate buffer, ph 7.4, varying concentrations of hydrogen peroxide (from 1 to 20 mm), and varying HPII hematin concentrations over the nanomolar range, at 25 C (see Methods). ND, not determined. tants are summarized in Table I. His128Ala and His- 128Asn gave no spectral changes upon H 2 O 2 addition, but Asn201 mutants typically showed a decrease in

5 62 OBINGER ET AL. FIG. 3. Kinetics of the reactions of wild-type HPII and N201D mutant catalases with hydrogen peroxide. (A) Reaction of HPII wild type (1.6 mm) with (a) 15 mm, (b) 35 mm, (c) 125 mm, and (d) 250 mm H 2 O 2. (B) Reaction of N201D mutant (1.7 mm) with (a) 15 mm, (b) 35 mm, and (c) 250 mm H 2 O 2. The medium contained 50 mm potassium phosphate buffer, ph 7.4, 25 C. Time courses at 405 nm were monitored in the stopped-flow system (see Methods). Catalatic Activity Catalatic activity of HPII (wild type), measured at 240 nm (Methods), was proportional to enzyme (hematin) concentration with an overall rate constant of M 01 s 01 at low [H 2 O 2 ] levels, Ç15% of the beef liver catalase activity. At high peroxide levels the reac- tion rate lost proportionality to [H 2 O 2 ] and a k cat value of about 27,000 s 01 was calculated. The apparent K m was therefore Ç20 mm H 2 O 2 (see Discussion). Some of the kinetic constants for HPII and its mu- tants are contained in Table I. His128Ala and His- 128Asn are catalytically inactive, whereas mutation of Asn201 gave partially active forms (Table I). Both wild type and mutants are sensitive to hydrogen peroxide concentrations higher than 100 mm. At these concen- trations there is a progressive inhibition of the enzyme (not shown), although spectral shifts were not detect- able. The rates of compound I formation of HPII and its Asn201 mutants were measured directly by stoppedflow spectroscopy. A typical series of time traces in the Soret region at various concentrations of hydrogen peroxide is shown in Fig. 3A; Fig. 3B shows the same experiment with the N201D mutant. In all cases the time courses were fitted as single exponential functions and apparent second-order rate constants calculated by plots of k obs vs [H 2 O 2 ]. Both for wild-type HPII and for its Asn201 mutants k 1 (app) values were reduced approximately in proportion to the reduction in overall rates (Table I). As Asn201Ala, Asn201Asp, and Asn201Gln also showed similar hypochromicities in the Soret region upon H 2 O 2 addition to that of HPII wild type (Table I), we conclude that k 2 (app) values for the mutants decrease to the same extent as do k 1 (app) values (Eqs. [1] [4]). Peroxidatic Activity Figure 4 shows a typical series of time traces of the reaction of HPII compound I with formate at 405 nm, using the sequential stopped-flow method (Methods). Similar traces were obtained with ethanol as hydrogen donor (not shown). The kinetics are of first-order pro- cesses whose calculated pseudo-first-order rate constants are proportional to donor concentration (Table II). Both ethanol and formate oxidations by WT compound I are much slower than the corresponding reactions with beef liver enzyme. The same experiments performed with Asn201 mutants again indicated a prominent role for asparagine in orienting the electron donor. All these mutants react more slowly than does the wild-type enzyme (Table II). Absence of Compound II in HPII Catalase WT and mutant enzymes never showed any shifts in their spectra which would indicate compound II formation. Neither potassium ferrocyanide, a one-electron re- ductant of compound I in beef liver catalase [discussed by Deisseroth and Dounce (28)], used in a concentration range from 10 mm to 10 mm, nor incubation with milli- molar ascorbate ever produced a species similar to the secondary peroxide compound of beef liver catalase (re- sults not shown). Heme Conversion HPII N201H contained protoheme and showed no more than 1% of the activity of HPII wild type. Its

6 HEME d SYNTHESIS IN Escherichia coli HPII CATALASE 63 FIG. 4. Sequential traces for the reaction of HPII wild-type catalase compound I with formate. HPII wild-type (1.6 mm heme) compound I was formed with H 2 O 2 as in Fig. 3 and then reacted with (a) 0.1 mm, (b) 1 mm, and (c) 2 mm sodium formate. Compound I formation occurred within the first few milliseconds. The medium contained 50 mm potassium phosphate buffer, ph 7.4. Time courses at 405 nm were monitored in the stopped-flow system at 25 C. spectrum was shifted to that of a heme d species in the pseudo-first-order kinetics for H 2 O 2 dismutation, the presence of hydrogen peroxide or hydrogen peroxidegenerating rate constant for H 2 O 2 degradation by N201H increases systems such as ascorbate plus molecular with time; this indicates that the enzyme was con- oxygen (cf. Ref. 18). This conversion is accompanied by verted during the reaction course from a partially ac- an increase in catalytic activity, as shown in Table I. tive to a more active form. Figures 5A and 5B compare the overall kinetics of During the first few minutes the Soret band of hydrogen peroxide degradation by WT enzyme with the N201H during its reaction with hydrogen peroxide (and kinetics of the mutants N201D and N201H. In contrast also some peaks in the visible range) had approxi- to the WT and N201D catalases, which exhibit typical mately 10 14% less intensity than in the resting state. This suggested formation of a protoheme compound I which slowly shifts to the corresponding heme d form TABLE II (Fig. 6). The protoheme form, with peaks at 403, 500, Transient State Rate Constants for the Reactions of HPII 538, and 632 nm, was transformed to a heme d species Wild-Type and Mutant Catalases and Peroxide Compounds with a Soret band at 407 nm and a major peak in the with Hydrogen Peroxide and with Peroxidatic Donors visible region at 578 nm. As shown previously (20), the cyanide spectra of these two different heme forms also k 1 (app) k 2 (app) Formate Ethanol characterize them as protoheme low-spin and heme d Catalase (M 01 s 01 ) (M 01 s 01 ) (M 01 s 01 ) (M 01 s 01 ) low-spin complexes, respectively. HPII wild type N201 A DISCUSSION N201D E. coli HPII contains heme d (the lactone form is N201 Q ND ND shown in Fig. 7A) instead of protoheme, but in an analogous pocket at a similar depth in the protein to the Note. Conditions: 50 mm potassium phosphate buffer, ph 7.4, and 25 C (see Methods and Table 1). k 1 (app), the rate constant for reaction of ferric enzyme with peroxide; k 2 (app), the rate constant for cally in Fig. 7B. The present paper describes some ef- heme in beef liver enzyme (16), as illustrated schematireaction of compound I with peroxide; and k 2 (app), the rate con- fects of single-point mutations of residues on the distal stant(s) for reaction of compound I with other hydrogen donors (forside of the heme, analogs of which are known to be mate and ethanol) were all calculated from Eqs. [1] and [4] using the steady-state compound I concentrations given in Table I. ND, important for dismutation of H 2 O 2 by beef liver catalase not determined (very low activity). (19). Replacement of the distal histidine with alanine

7 64 OBINGER ET AL. FIG. 5. Hydrogen peroxide degradation by mutant enzymes. (A) N201H kinetics. Catalytic decay of 45 mm H 2 O 2 catalyzed by (a) 73 nm and (b) 220 nm HPII N201H mutant enzyme at ph 7.4, 25 C. Measurements were carried out at 240 nm. (B) N201D kinetics. Catalytic decay of (a) 4.8 mm H 2 O 2, (b) 9.5 mm H 2 O 2, (c) 19.5 mm H 2 O 2, and (d) 47.6 mm H 2 O 2 catalyzed by the N201D mutant enzyme. Reactions were started by addition of 19 nm (hematin) catalase. Other conditions as in (A) and Methods. or asparagine results in mutant HPII enzymes H128A v Å 2k 1k 2 es and H128N which contain only protoheme, whereas Å k es [5] {k 1 / k 2 } replacement of N201A, N201Q, and N201D all contain heme d species. One distal asparagine mutant, N201H, Values of k typically fall between 0.6 and contains protoheme as isolated (13, 18). M 01 s 01 (29). With a k value of M 01 s 01 HPII The overall reaction rate constant (k ) for various has about 15% the activity of beef liver catalase, and catalases is given by Eq. [5], where k 1 and k 2 are the both k 1 and k 2 are reduced in magnitude, preserving rates of formation and reduction of compound I (see some 30 40% of compound I in the steady state. Equation Eqs. [2] and [3] above). [5] is valid for beef liver catalase up to molar levels

8 HEME d SYNTHESIS IN Escherichia coli HPII CATALASE 65 FIG. 6. Heme conversion of the HPII N201H mutant enzyme by hydrogen peroxide and ascorbate. N201H mutant enzyme (2.5 mm hematin) was incubated with 50 mm hydrogen peroxide in the presence of 1 mm ascorbate. Addition of H 2 O 2 plus ascorbate to ferricatalase (a) induces a steady-state protoheme compound I spectrum (b) which is visible for 5 to 10 min. Finally the bands in the Soret and visible region shift (c), forming a species indicating either heme d itself or a precursor form. The medium was 50 mm potassium phosphate buffer, ph 7.4, at 25 C. of peroxide [an apparent K m for H 2 O 2 of 1.1 M was re- tween the His74 and Asn147 residues (cf. Fig. 7B) before ported at 20 C by Ogura (30), corresponding to a k cat of interacting with the heme iron. These two residues 10 7 s 01 ], but HPII apparently has a much lower maxi- preorient the substrate and then interact with it via mal turnover (27,000 s 01 ) and a correspondingly hydrogen bonding (15). Replacement of the critical his- smaller K m for hydrogen peroxide (Ç20 mm). tidine ligand in HPII gave (protoheme) enzymes which The initial step in the formation of compound I is the were catalytically quite inactive. His128 may be essenentrance of H 2 O 2 into the distal side of the heme crev- tial for proton abstraction from the peroxide as well as ice. The peroxide is sterically constrained to move be- for transfer of hydrogen to the distal peroxide oxygen FIG. 7. Proposed structure of heme d and its location in the HPII active site. (A) Lactone or spiro form of heme d (16). Compare with the structures postulated by Timkovich and Bondoc (35). (B) Proposed arrangement of residues in the heme pocket, derived from the HPII crystal structure of Bravo et al. (16) (obtained from a Quanta image visualized with a Silicon Graphics Iris Unix-based computer).

9 66 OBINGER ET AL. during O O cleavage, when a molecule of water is cally in the initial steady state. This is followed by formed. conversion to a heme d species and formation of a more Changing the possible essential asparagine residue active enzyme. The details of this heme modification (N201) yielded mutants ranging in overall activity from for HPII are still unknown. A serine residue (located 24% to nearly undetectable compared to the wild type. Ç2 Åbelow pyrrole ring III; see Fig. 7B) substitutes Although the asparagine thus may not be essential, it for an alanine in beef liver catalase and may assist in may nonetheless act via hydrogen bonding to orient the hydroxylation to form heme d (cf. Ref. 16). hydrogen peroxide. As N201Q showed almost no activity, whereas N201A had Ç24% and N201D Ç10% of ACKNOWLEDGMENTS wild type, N201 has an obvious stereochemical role, This work was supported by Canadian NSERC Research Grant A- incorporation of an extra CH 2 -group hindering peroxide 0412 to P.N., by the Jubilaeumsfond der Oesterreichischen Nationbinding to the active site. albank (P4828) to C.O., and by an NSERC grant to P.C.L. at the In contrast to beef liver catalase where in steady University of Manitoba. Preliminary experiments were carried out state only 25 30% of the hemes are in the compound by Alex Hillar, now of the University of Manitoba, as part of his M.Sc. project at Brock University. We thank Ms. Brenda Tattrie for I form during the steady state (31), HPII has at least skilled technical assistance at Brock University and Jacek Switala 40% of the enzyme in this form (Fig. 2). This percentage of the University of Manitoba for the preparation of mutant organremains the same for mutants showing much lower isms and the associated enzymes. Parts of this work are contained overall activity than the wild-type enzyme. Both rate in the M.Sc. thesis of M.M. at Brock University. constants k 1 and k 2 (Eqs. [2] and [3]) are therefore affected by distal residue modifications. REFERENCES The observed differences in activity compared with 1. Yonetani, T., and Schleyer, H. (1967) J. Biol. Chem. 267, 8827 beef liver catalase may be correlated with differences not only in the heme pocket (Fig. 7) but also involving 2. Nicholls, P. (1962) Biochim. Biophys. Acta 60, the peroxide entry channel. This channel leading 3. Gibson, J. F., Ingram, D. J. E., and Nicholls, P. (1958) Nature from the exterior to heme d contains unique phenylala- 181, nine and tryptophan residues (16) that may occlude the 4. Catalano, C. E., Choe, Y. S., and Ortiz de Montellano, P. R. diffusion pathway. Thus, ethanol and formate oxidations (1989) J. Biol. Chem. 264, by HPII compound I are much slower than with 5. Davies, M. J. (1991) Biochim. Biophys. Acta 1077, beef liver catalase. The reaction of HPII with cyanide 6. Davies, M. J., and Puppo, A. (1992) Biochem. J. 281, as a heme ligand is also slow (20). 7. Ortiz de Montellano, P. R. (1992) Annu. Rev. Pharmacol. Toxicol. Beef liver catalase has two phases of catalatic activ- 32, ity, termed a and b by George (32). In the b state the 8. Miller, M. A., Bandyopadhyay, D., Mauro, J. M., Traylor, T. G., enzyme exists in a less active form, the rate and extent and Kraut, J. (1992) Biochemistry 31, of inhibition being dependent upon peroxide concentra- 9. Dunford, H. B. (1991) in Peroxidases in Chemistry and Biology (Everse, J., and Grisham, M. B., Eds.), Vol. II, pp , CRC tion (32, 33). Chance interpreted the change in terms Press, Boca Raton, FL. of compound II formation (34) and this analysis was 10. Dolphin, D., Forman, A., Borg, D. C., Fajer, B., and Felton, R. H. extended by Nicholls and Schonbaum (24). E. coli HPII (1971) Proc. Natl. Acad. Sci. USA 68, compound I, on the other hand, reacts with peroxides 11. Nicholls, P. (1964) Biochem. Biophys. Acta 81, exclusively in a two-electron process. No compound II 12. Kirkman, H. N., and Gaetani, G. F. (1984) Proc. Natl. Acad. Sci. or compound III formation has been seen. HPII also USA 81, does not bind NADPH, whose role is thought to be pre- 13. Hillar, A., Nicholls, P., Switala, J., and Loewen, P. C. (1994) vention of compound II formation and inactivation by Biochem. J. 300, H 2 O 2 (13, 14). Nevertheless, HPII also shows inhibition 14. Kirkman, H. N., Galiano, S., and Gaetani, G. F. (1987) J. Biol. at high peroxide levels, but this inhibition does not Chem. 262, appear to involve compound II formation. Its mecha- 15. Fita, I., and Rossman, M. G. (1985) J. Mol. Biol. 185, nism is unclear and it may not be reversible. 16. Bravo, J., Verdaquer, N., Tormo, J., Betzel, C., Switala, J., and It has been suggested by Timkovich and Bondoc (35) Loewen, P. (1995) Structure 3, that protoheme is first bound to HPII apoenzyme and 17. Loewen, P. C., and Switala, J. (1986) Biochem. Cell. Biol. 64, that heme hydroxylation is catalyzed by HPII itself Loewen, P. C., Switala, J., von Ossowski, I., Hillar, A., Christie, utilizing one of the first H 2 O 2 molecules to react catalyt- A., Tattrie, B., and Nicholls, P. (1993) Biochemistry 32, ically. A role for His128 in the conversion has been demonstrated by Loewen et al. (18), but Asn201 is not 19. Murthy, M. R. N., Reid, T. J., Sicignano, A., Tanaka, N., and an absolute requirement. Rossmann, M. G. (1981) J. Mol. Biol. 152, Heme d formation from the protoheme species of 20. Maj, M., Nicholls, P., Obinger, C., Hillar, A., and Loewen, P. N201H occurs during the first few turnovers via protoheme (1996) Biochim. Biophys. Acta 1298, compound I, which can be observed spectroscopi- 21. Dawson, J. H., Bracete, A. M., Huff, A. M., Kadkhodayan, S.,

10 HEME d SYNTHESIS IN Escherichia coli HPII CATALASE 67 Zeitier, C. M., Sono, M., Chang, C. K., and Loewen, P. C. (1992) 28. Deisseroth, A., and Dounce, A. L. (1970) Physiol. Rev. 50, 319 FEBS Lett. 295, Beers, R. F., and Sizer, I. W. (1952) J. Biol. Chem. 195, Lardinois, O. M. (1995) Free Radical Res. 22, Ogura, Y. (1955) Arch. Biochem. Biophys. 57, Nelson, D. P., and Kiesow, L. A. (1972) Anal. Biochem. 49, Chance, B., Sies, H., and Boveris, A. (1979) Physiol. Rev. 59, Nicholls, P., and Schonbaum, G. R. (1963) in The Enzymes (Boyer, P., Lardy, H., and Myrback, K., Eds.), Vol. VIII, 2nd ed., 32. George, P. (1947) Nature 160, pp , Academic Press, San Diego. 33. George, P. (1949) Biochem. J. 44, Chance, B. (1952) J. Biol. Chem. 194, Chance, B. (1950) Biochem. J. 46, Keilin, D., and Hartree, E. F. (1945) Biochem. J. 39, Timkovich, R., and Bondoc, L. L. (1990) Adv. Biophys. Chem. 1, 27. Keilin, D., and Hartree, E. F. (1955) Biochem. J. 60,